Even the prettiest faces are built using junk. In mice, the shapes of the face and skull are finely tuned by junk DNA, so called because it was initially thought to lack function since it doesn't encode proteins. The same junk DNA sequences are found in humans, so they are probably also shaping our faces.

This finding could help us make sense of some congenital conditions, such as cleft palates, that can develop even when the genes that shape the face appear to be working normally.

There is a huge degree of variation in human faces but, as family resemblances show, the overall shape is heavily constrained by genetics. However, so far, geneticists have identified only a small number of genes that influence the shape. These explain just a tiny fraction of the variation seen in human faces.

According to Axel Visel of the Lawrence Berkeley National Laboratory in California and his colleagues, more variation is controlled by distant-acting enhancers. These are short sequences of DNA, in non-coding regions of the genome, that can influence the activity of the facial genes, even if they are a long way along the DNA strand.

"Enhancers are part of the 98 per cent of the human genome that is non-coding DNA – long thought of as 'junk DNA'," says Visel. "It's increasingly clear that important functions are embedded in this 'junk'."
Face enhancers

Visel and his colleagues used a technique called optical projection tomography, which allows them to build up a three-dimensional model of a developing mouse embryo and show how gene expression varies in each region. This revealed some 120 enhancers that were active in various regions of the developing face. To work out what the enhancers were doing, the team chose three and engineered three groups of mice to each lack one of them.

When the mice were 8 weeks old, the team compared their skulls and faces to those of a control group of mice with all enhancers intact. They found that each enhancer had a subtle effect on face shape. For instance, deleting one enhancer left mice with faces that were longer – but skulls that were broader and shorter – than the control mice.

It's an important study, says Lavinia Paternoster, a geneticist at the University of Bristol, UK, because it identifies the key areas of the genome for face shape. "Studies such as this will enable us to focus on regions of the genome that are more likely to harbour important genetic variants and hence might mean that we can identify these variants with smaller sample sizes than is usually necessary."

However, she thinks the junk DNA might not be as important as Visel suggests. She has performed one of the few studies so far to identify genes involved in facial morphology. She investigated the entire genome – including coding and non-coding regions – of thousands of individuals, and found little evidence that non-coding regions have a powerful effect on face shape. "I am in no doubt that some enhancers do harbour some genetic variants that will influence face shape," she says. "But they are not the Holy Grail of missing heritability."
Cleaving palates

None of these effects on face shape were dramatic enough to, say, cause a cleft palate. But several enhancers can act on a single gene, says Visel. So if some or all of these enhancers carry mutations, their cumulative effect might lead to such dramatic facial changes.

"There are many cases of craniofacial pathologies – including a significant number of cases of clefts of the lip or palate – that cannot currently be explained by mutations in protein-coding genes," says Visel. "Based on our studies in the mouse model, it is possible or even likely that in some of these cases mutations in enhancers play a role."

So perhaps rather than looking for mutations in the genes, we should be focusing on mutations in the enhancers that influence those genes.

"This will certainly help us to understand the underlying causes of these defects better, and will eventually help in the diagnosis, possibly also prevention or treatment of such conditions," says Visel.